Initial Assessment of the Molecular Epidemiology of bla_NDM-1 in Colombia

Abstract

We report complete genome sequences of four bla_NDM-1-harboring Gram-negative multidrug-resistant (MDR) isolates from Colombia. The bla_NDM-1 genes were located on 193-kb Inc FIA, 178-kb Inc A/C2, and 47-kb (unknown Inc type) plasmids. Multilocus sequence typing (MLST) revealed that these isolates belong to sequence type 10 (ST10) (Escherichia coli), ST392 (Klebsiella pneumoniae), and ST322 and ST464 (Acinetobacter baumannii and Acinetobacter nosocomialis, respectively). Our analysis identified that the Inc A/C2 plasmid in E. coli contained a novel complex transposon (Tn125 and Tn5393 with three copies of bla_NDM-1) and a recombination “hot spot” for the acquisition of new resistance determinants.

TEXT

At this time, bla_NDM is recognized as a major global health threat. Guatemala and Colombia reported the first cases of bla_NDM-1-harboring isolates in Latin America (1, 2). In both instances, bla_NDM-1 was discovered in hospital-acquired, clonally related Klebsiella pneumoniae isolates that were recovered from patients who had not travelled recently. The molecular epidemiology of bla_NDM carbapenemases in South America has been investigated only in a limited fashion, and data in Colombia are very scarce. In order to understand the dissemination of bla_NDM-1 in Colombia, we performed a genomic analysis of four sentinel isolates, Acinetobacter baumannii, Acinetobacter nosocomialis, Escherichia coli, and K. pneumoniae collected in 2012, shortly after the first reported outbreak (1).

(Part of this work was presented at the 55th Annual Interscience Conference on Antimicrobial Agents and Chemotherapy [ICAAC], San Diego, CA, 17 to 21 September 2015.)

Phenotypic characterization revealed that two of the isolates (E. coli and A. nosocomialis) were resistant to all antibiotics tested, including polymyxin B and tigecycline (see Table S1 in the supplemental material). On the other hand, the A. baumannii isolate was also multidrug resistant but susceptible to polymyxin B, tigecycline, and ciprofloxacin. The K. pneumoniae isolate was susceptible to polymyxin B and exhibited relatively low susceptibility to carbapenems, highlighting once again the difficulty clinical microbiology laboratories have detecting carbapenemase genes that are expressed at low levels (3, 4). None of the isolates tested positive for carbapenemases using the modified Hodge test, but they tested positive (with the exception of the E. coli isolate) in the three-dimensional (3D) bioassay using an imipenem disk (5). Double-disk synergy testing using EDTA (DDST+EDTA) confirmed the presence of a metallo-β-lactamase for the members of the family Enterobacteriaceae, but not in the Acinetobacter species isolates.

In order to understand the genetic background of these early bla_NDM-1-containing strains, the complete chromosome and plasmid sequences were obtained by assembly of Pacific Biosciences single-molecule real-time (SMRT) sequence data, with the exception of K. pneumoniae, where the chromosome was assembled into three ordered contigs (Table 1). Genome sequencing results showed that all isolates possessed multiple plasmids (Table 1) and revealed that bla_NDM-1 was localized in one plasmid per strain, as confirmed by S1 nuclease pulsed-field gel electrophoresis (S1-PFGE) (6).

TABLE 1.

Accession number and resistome of bla_NDM-1-harboring isolates

Species	Parameter	Chromosome	Plasmids
A. baumannii	Accession no.	NZ_CP010397.1	CP010398.1	CP010399.1	CP010400.1
	Size (bp)	3,902,527	114,848	47,274	9,327
	Resistance determinant(s)	blaADC-80 and blaOXA-94	None	aph(3′)VIIa and blaNDM-1	None
A. nosocomialis	Accession no.	CP010368.1	NZ_CP010369.1	CP010903.1	CP010370.2
	Size (bp)	3,858,956	89,111	66,409	47,274
	Resistance determinant(s)	blaADC-80	None	None	aph(3′)VIIa and blaNDM-1
K. pneumoniae	Accession no.	NZ_JWRK01000001.1	CP010390.1	CP010391.1
	Size (bp)	5,329,244a	198,371	178,193
	Resistance determinant(s)	blaCTXM-15, blaSHV-11, oqxA, oqxB, and fosA	strA, strB, aac(3′)IIa, aac(6′)lb-a, qnrB66, sul2, tetA, dfrA14, catB3, blaTEM-1, blaCTXM-15, and blaOXA-1	aph3′VIa, aacA29, aadA2, mph(E), msr(E), catB3, cmlA1, sul2, sul1, blaNDM-1, and blaCARB-2
E. coli	Accession no.	NZ_CP010371.1	NZ_CP010373.2	NZ_CP010372.1
	Size (bp)	4,761,012	193,908	151,583
	Resistance determinant(s)	sul1	strA, strB, catA1, sul2, sul1, tetB, dfrA1, aadA16, and blaNDM-1 (3 times)	catA1, sul1, tetB, dfrA7, and blaTEM-1

^aThe genome was not closed. The size is estimated based on the size of the three contigs.

Multilocus sequence type (MLST) analysis revealed that Acinetobacter species isolates belonged to sequence type 322 (ST322) (A. baumannii) and ST464 (A. nosocomialis), none regarded as “high-risk” clones (7). Both A. baumannii and A. nosocomialis harbored three plasmids and carried the bla_NDM-1 gene on a Tn125 backbone (Fig. 1) located on a 47,274-bp plasmid that was 99% similar to plasmid pNDM-BJ01 (GenBank accession no. JQ001791.1) reported in an Acinetobacter lwoffi isolate from China (8, 9). This plasmid also carried the aminoglycoside phosphotransferase aph(3′)VIIa gene, and a type IV secretion system (T4SS) gene cluster encoding a P-type T4SS that has been reported to encode a short, rigid pilus characteristic of broad-host-range conjugative plasmids (10).

FIG 1.

Organization of the bla_NDM-1-containing element in E. coli. The first two copies of bla_NDM-1 are contained within the Tn125 transposon (flanked by ISAba125 [shown in black]). The third copy lacks the right-flanking ISAba125 and is followed instead by a Tn5393-like element. Multiple transposable elements are found within this region, indicating that it may serve as a “hot spot” for the incorporation of new resistance determinants through homologous recombination via IS elements, site-specific recombination, or transposition.

We next discovered that a chromosomally encoded class C β-lactamase bla_ADC-80 was found in both Acinetobacter species isolates. This surprising finding is particularly interesting, as it highlights the possibility that bla_ADC evolved similarly in two different species of Acinetobacter. Consistent with the species identification, A. baumannii also harbored bla_OXA-94 (a bla_OXA-51-like derivative). None of the other plasmids contained additional resistance genes, and interestingly, A. baumannii resistance islands (AbaRI) were not found in the chromosome or on plasmids harbored by either of the Acinetobacter species isolates.

In contrast with Acinetobacter species isolates, both members of the Enterobacteriaceae contained multiple resistance genes in two large plasmids (151 to 198 kb) (Table 1), including bla_CTXM-15, consistent with the previously documented predominance of that extended-spectrum β-lactamase (ESBL) in Colombia (11, 12). The K. pneumoniae isolate belonged to ST392, previously associated with the dissemination of bla_KPC, bla_OXA-48, and other ESBLs (13, 14). In addition, this isolate also harbored a chromosomal bla_SHV-11 and bla_CTX-M-15, the latter located downstream of ISEcp1, as previously reported in other Enterobacteriaceae isolates from Spain, Japan, Germany, the Netherlands, and the United Kingdom (15–18). The largest plasmid of K. pneumoniae (198 kb) was a multireplicon Inc FII/FIB type plasmid, and it carried antibiotic determinants conferring resistance to aminoglycosides [aph(3′)-Ib, aph(6′)-Id, aac(3)-IIa, and aac(6′)-Ib-cr], quinolones (qnrB66), sulfonamides (sul2), tetracycline [tet(A)], trimethoprim (dfrA14), chloramphenicol (catB3), and β-lactams (bla_TEM-1, bla_CTX-M-15, and bla_OXA-1). In this case, the second copy of bla_CTX-M-15 was found to be part of a previously reported structure: a Tn3-like transposon also carrying bla_TEM-1 which has its tnpA gene disrupted by ISEcp1-bla_CTX-M-15 (19). In this plasmid, this entire structure is also followed by an IS26 element, previously shown to have a critical role in the mobilization and reorganization of antibiotic resistance genes in Gram-negative bacteria (20, 21).

The bla_NDM-1-bearing plasmid (178 kb) contained an Inc A/C2 replicon, extensively associated with antibiotic resistance in Gram-negative bacteria (22). The plasmid backbone shares similarity with other plasmids carrying bla_NDM-1 and other β-lactamases in a variety of Gram-negative species (see Table S2 in the supplemental material). Additionally, this plasmid carried determinants conferring resistance to most antimicrobial classes, including β-lactams (bla_CARB-2), aminoglycosides [aph(3′)-VIa, aacA29, and aadA2], chloramphenicol (catB3 and cmlA1), sulfonamides (sul2 and sul1), macrolides [mph(E)], streptogramin B (strB), and lincosamide [msr(E)].

The E. coli isolate belonged to ST10 (phylogroup A), which has been associated with ESBLs and hyperexpressed AmpC enzymes (7). E. coli harbored most of the resistance determinants in plasmids; only the sulfonamide resistance gene sul2 was present in the chromosome. The Inc FIA/FIB 151-kb plasmid carried bla_TEM-1, catA1, sul1, tetB, and dfrA7, while the 193-kb Inc A/C2 plasmid harbored not only bla_NDM-1 but also strA. strB, catA1, sul2, sul1, tetB, and dfrA1. It is noteworthy that there were three tandem repeats of bla_NDM-1 in the 193-kb plasmid, two of them within a Tn125 structure and the last one lacking the right side copy of ISAba125 (Aba stands for A. baumannii) (Fig. 1). We interpret this to be a consequence of insertion of Tn125 within a Tn5393-like structure, as evidenced by the presence of tnpA, tnpR, strA, and strB, characteristic of this Tn3 transposon, originally reported for Erwinia amylovora (23), but now found in several Gram-negative species in clinical, ecological, and agricultural niches (24–27). This complex array of transposons is followed by an aminoglycoside resistance gene (aadA16) flanked by transposable genetic elements, indicating that this whole region could be serving as a “hot spot” for the incorporation of genetic determinants either by homologous recombination via IS elements, site-specific recombination, or transposition. A similar 20-kb resistance region is found on environmental plasmid pRSB101 which was originally isolated from bacterial populations residing in the activated sludge compartment of a wastewater treatment plant (28). Furthermore, one of those transposable elements was identified as IS26 and found not only flanking the above-mentioned region, but also next to the first copy of bla_NDM-1 containing Tn125. This would reinforce the hypothesis of a “hot spot region,” given the replicative transposition mechanism of IS26 and its previously shown critical role in the mobilization and reorganization of antibiotic resistance genes in Gram-negative bacteria (20, 21). Most importantly, although bla_NDM in tandem repeats has been observed before, this is to our knowledge the first report of such a structure in E. coli. In both previously reported cases, it occurred in K. pneumoniae isolates: the first from a Taiwanese patient with a hospitalization history in New Delhi, India (250-kb Inc FIB/FII type plasmid) (29) and the second from an outbreak in a neonatal unit in Nepal (304-kb Inc HIB/FIB type plasmid) (30).

All strains were nosocomially acquired and isolated from elderly patients with severe systemic infections, three patients, who presented several comorbidities, died (see Table S3 in the supplemental material). First, since evidence of international travel or travel to Bogota, Colombia (where the first Colombian bla_NDM-1 was reported) could not be established for any of the patients or their families and second, given that they originally lived in rural areas or small cities, this emergence in a variety of species in two different geographic locations, is extremely worrisome. Colombia has often been among the first countries in the region to report the circulation of important resistance determinants, including CTX-M-15, KPC, and NMC-A, all of which have become widely disseminated, even becoming endemic, as is the case for KPC (31–34). Even though information regarding molecular epidemiology of bla_NDM-1 in Colombia is still very limited, the National Institute of Health of Colombia is reporting increased numbers of patients infected with NDM-producing bacteria. Interestingly, K. pneumoniae and Providencia rettgeri are the most prevalent bla_NDM-1-expressing Gram-negative bacterial species (35, 36). We hypothesize that the rapid spread of this resistance gene (bla_NDM-1) is aided by the circulation of broad-host-range, transferable plasmids such as Inc A/C found in this study.

The widespread dissemination of bla_NDM in Colombia portends a significant antibiotic resistance problem in Latin America (1). Colombia's situation may be only the “tip of the iceberg”; therefore, studies assessing the real prevalence of bla_NDM, especially in countries where few reports are available, are warranted. It is of great importance that the findings of surveillance and genomic studies like the present one help inform new and more-effective infection control and stewardship programs that can be translated into appropriate national policies to prevent a situation where it becomes endemic.

Sequence accession numbers.

Sequences have been deposited in GenBank under the accession numbers given in Table 1.

Funding Statement

Research reported in this publication was supported by the Genome Center for Infectious Diseases grant U19AI110819 to M.D.A. Research reported in this publication was supported in part by the National Institute of Allergy and Infectious Diseases of the National Institutes of Health under award numbers R01AI100560, R01AI063517, and R01AI072219 to R.A.B. This study was supported in part by funds and/or facilities provided by the Cleveland Department of Veterans Affairs, award number 1I01BX001974 to R.A.B. from the Biomedical Laboratory Research & Development Service of the VA Office of Research and Development, and the Geriatric Research Education and Clinical Center VISN 10 to R.A.B.

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health or the Department of Veterans Affairs.